I have confirmed this in production,
with the default max-entries.
I have a bucket that I'm no longer writing to. Radosgw-agent had
stopped replicating this bucket. radosgw-admin bucket stats shows
that the slave is missing ~600k objects.
I uploaded a 1 byte file to the bucket. On the next pass,
radosgw-agent replicated 1000 entries.
I'm uploading and deleting the same file every 5 minutes. I'm
using more inter-colo bandwidth now. This bucket is catching up,
slowly.
For now, I'm going to graph the delta of the total number of
objects in both clusters. If the slave is higher, it's catching
up. If it's lower, it's falling behind.
On 2/6/14 18:32 , Craig Lewis wrote:
On 2/4/14 17:06 , Craig Lewis wrote:
Now that I've started seeing missing objects, I'm not able to
download objects that should be on the slave if replication is
up to date. Either it's not up to date, or it's skipping
objects every pass.
Using my --max-entries fix (https://github.com/ceph/radosgw-agent/pull/8),
I think I see what's happening.
Shut down replication
Upload 6 objects to an empty bucket on the master:
2014-02-07 02:02 10k dc5674336e2212a0819b7abcb811e323
s3://bucket1/test0.jpg
2014-02-07 02:02 10k
dc5674336e2212a0819b7abcb811e323 s3://bucket1/test1.jpg
2014-02-07 02:02 10k
dc5674336e2212a0819b7abcb811e323 s3://bucket1/test2.jpg
2014-02-07 02:02 10k
dc5674336e2212a0819b7abcb811e323 s3://bucket1/test3.jpg
2014-02-07 02:03 10k
dc5674336e2212a0819b7abcb811e323 s3://bucket1/test4.jpg
2014-02-07 02:03 10k
dc5674336e2212a0819b7abcb811e323 s3://bucket1/test5.jpg
None show on the slave, because replication is down.
Start radosgw-agent --max-entries=2 (1 doesn't seem to replicate
anything)
Check contents of slave after pass #1:
2014-02-07 02:02 10k dc5674336e2212a0819b7abcb811e323
s3://bucket1/test0.jpg
Check contents of slave after pass #10:
2014-02-07 02:02 10k dc5674336e2212a0819b7abcb811e323
s3://bucket1/test0.jpg
Leave replication running
Upload 1 object, test6.jpg, to the master. Check the master:
2014-02-07 02:02 10k dc5674336e2212a0819b7abcb811e323
s3://bucket1/test0.jpg
2014-02-07 02:02 10k
dc5674336e2212a0819b7abcb811e323 s3://bucket1/test1.jpg
2014-02-07 02:02 10k
dc5674336e2212a0819b7abcb811e323 s3://bucket1/test2.jpg
2014-02-07 02:02 10k
dc5674336e2212a0819b7abcb811e323 s3://bucket1/test3.jpg
2014-02-07 02:03 10k
dc5674336e2212a0819b7abcb811e323 s3://bucket1/test4.jpg
2014-02-07 02:03 10k
dc5674336e2212a0819b7abcb811e323 s3://bucket1/test5.jpg
2014-02-07 02:06 10k
dc5674336e2212a0819b7abcb811e323 s3://bucket1/test6.jpg
Check contents of slave after next pass:
2014-02-07 02:02 10k dc5674336e2212a0819b7abcb811e323
s3://bucket1/test0.jpg
2014-02-07 02:02 10k
dc5674336e2212a0819b7abcb811e323 s3://bucket1/test1.jpg
Upload another file, test7.jpg, to the master:
2014-02-07 02:02 10k dc5674336e2212a0819b7abcb811e323
s3://bucket1/test0.jpg
2014-02-07 02:02 10k
dc5674336e2212a0819b7abcb811e323 s3://bucket1/test1.jpg
2014-02-07 02:02 10k
dc5674336e2212a0819b7abcb811e323 s3://bucket1/test2.jpg
2014-02-07 02:02 10k
dc5674336e2212a0819b7abcb811e323 s3://bucket1/test3.jpg
2014-02-07 02:03 10k
dc5674336e2212a0819b7abcb811e323 s3://bucket1/test4.jpg
2014-02-07 02:03 10k
dc5674336e2212a0819b7abcb811e323 s3://bucket1/test5.jpg
2014-02-07 02:06 10k
dc5674336e2212a0819b7abcb811e323 s3://bucket1/test6.jpg
2014-02-07 02:08 10k
dc5674336e2212a0819b7abcb811e323 s3://bucket1/test7.jpg
The slave doesn't get it this time:
2014-02-07 02:02 10k
dc5674336e2212a0819b7abcb811e323 s3://bucket1/test0.jpg
2014-02-07 02:02 10k
dc5674336e2212a0819b7abcb811e323 s3://bucket1/test1.jpg
Upload another file, test8.jpg, to the master:
2014-02-07 02:02 10k dc5674336e2212a0819b7abcb811e323
s3://bucket1/test0.jpg
2014-02-07 02:02 10k
dc5674336e2212a0819b7abcb811e323 s3://bucket1/test1.jpg
2014-02-07 02:02 10k
dc5674336e2212a0819b7abcb811e323 s3://bucket1/test2.jpg
2014-02-07 02:02 10k
dc5674336e2212a0819b7abcb811e323 s3://bucket1/test3.jpg
2014-02-07 02:03 10k
dc5674336e2212a0819b7abcb811e323 s3://bucket1/test4.jpg
2014-02-07 02:03 10k
dc5674336e2212a0819b7abcb811e323 s3://bucket1/test5.jpg
2014-02-07 02:06 10k
dc5674336e2212a0819b7abcb811e323 s3://bucket1/test6.jpg
2014-02-07 02:08 10k
dc5674336e2212a0819b7abcb811e323 s3://bucket1/test7.jpg
2014-02-07 02:10 10k
dc5674336e2212a0819b7abcb811e323 s3://bucket1/test8.jpg
The slave gets the 3rd file:
2014-02-07 02:02 10k dc5674336e2212a0819b7abcb811e323
s3://bucket1/test0.jpg
2014-02-07 02:02 10k
dc5674336e2212a0819b7abcb811e323 s3://bucket1/test1.jpg
2014-02-07 02:02 10k
dc5674336e2212a0819b7abcb811e323 s3://bucket1/test2.jpg
So I think the problem is caused by the shard marker being set to
the current marker after every pass, even if the bucket
replication caps on max-entries.
Updating the shard marker by uploading a file causes another pass
on the bucket, and the bucket marker is being tracked correctly.
I would prefer to track the shard marker better, but I don't see
any way to get the last shard marker given the last bucket entry.
If I track the shard marker correctly, then the stats I'm
generating are still somewhat useful (if incomplete). I'll be
able to see when replication falls behind because the graphs keep
growing.
The alternative is to change the bucket sync so that it loops
until it's replicated everything up to the shard marker. In this
case, I'll be able to see that replication is falling behind
because each pass takes longer and longer to complete.
What do you guys think?
Either way, I believe all my data is waiting to be replicated. I
just need to fix this issue, and upload another object to every
bucket that's behind.
_______________________________________________
ceph-users mailing list
ceph-users@xxxxxxxxxxxxxx
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
|
_______________________________________________
ceph-users mailing list
ceph-users@xxxxxxxxxxxxxx
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
